Compounded ketamine lozenges

For about five years, compounded ketamine troches and lozenges (small dissolving tablets made by specialty compounding pharmacies, taken at home) were the fastest-growing segment of mental health treatment in the country. Online clinics popped up, telehealth practices started prescribing them after a single Zoom visit, the boxes showed up in the mail, and patients took them at home. A lot of those patients got real benefit. A lot of them also got handed a controlled substance with no real oversight, and predictably, that started causing problems.

The FDA started cracking down in 2024 and 2025. Some of the largest telehealth ketamine companies shut down or got their compounding pharmacy relationships cut. Some are still operating in something close to a legal gray zone. Some have shifted to in-person clinic models. The landscape is mid-transition, which means the conversation about whether to use a compounded ketamine program in 2026 looks very different than it did even eighteen months ago.

What ketamine does and why people wanted lozenges

Ketamine has rapid-onset antidepressant effects through NMDA antagonism (NMDA is a glutamate receptor system in the brain, and blocking it for a few hours seems to kick off a cascade where the brain forms new neural connections quickly, which is the working theory for how ketamine pulls depression out within hours instead of weeks). IV ketamine in a clinic, dosed correctly, can shift a treatment-resistant depression (depression that hasn’t responded to at least two adequate antidepressant trials) within hours rather than the four to six weeks the standard medications need. Spravato (esketamine nasal spray, the FDA-approved nasal version dosed under medical supervision in a clinic) is the regulated equivalent with structured monitoring built in.

Compounded ketamine in oral form (lozenges, troches, sometimes nasal sprays) is the cheap at-home version. The bioavailability of oral ketamine is lower than IV, maybe 20 to 25 percent, which means more of what you take never makes it into your system in usable form. The pharmacokinetics are less predictable. The dissociation experience (the floaty, slightly-out-of-body sensation that’s part of how ketamine works) is less controlled at home than in a supervised setting. But it’s a fraction of the cost, doesn’t require a clinic visit, and worked for a meaningful percentage of patients who tried it.

Worth being honest about the evidence here, because the marketing rarely is. The intravenous and intranasal data are where most of the real research sits. Oral and sublingual ketamine has been studied far less, and the studies that exist are smaller and shakier. A 2021 Cochrane review of ketamine for depression found that ketamine may beat placebo in the first 24 hours, but rated that evidence as very low certainty and noted that how it translates into actual practice isn’t clear. The regulated nasal version, Spravato, has stronger data behind it than the compounded oral troches do, and anybody telling you the lozenge is just as proven is glossing over the part where it mostly isn’t.

The boom

From 2020 to 2024, online ketamine companies grew fast. Mindbloom, Better U, Joyous, others. Sign up online, brief telehealth visit, get a box in the mail, dose at home with a guide on your phone. Prices ranged from $250 to $700 a month depending on the company and the protocol.

Some patients did remarkably well, the kind of well where a depression that had been chewing on them for years lifted in a way nothing else had managed. Some had bad experiences, including the dissociation tipping into something genuinely unpleasant or scary. Some developed dependence patterns where the medication started feeling less like a treatment and more like something they needed every few days. And some used the dissociation for what was essentially recreational purposes, with the medical framing functioning as legal cover for the prescription. The variability of outcomes was wide enough that nobody could honestly defend the whole category without also acknowledging the rougher cases.

The clinical concern across most of the at-home programs was that there was often no real psychiatric follow-up, no integration of the experiences with talk therapy or any other supportive framework, no real titration based on response, and no monitoring for the things that genuinely need monitoring (urinary tract problems with long-term ketamine use, escalating dissociative tendencies, frequency-of-use patterns drifting toward dependence).

The crackdown

FDA started enforcement actions against several of the larger compounding pharmacies in 2024 and 2025, mostly aimed at mass-produced compounded products that were essentially functioning as unapproved drug manufacturing under a compounding-pharmacy label. The DEA also got more involved with the controlled-substance aspect, since ketamine is a Schedule III drug and the at-home dosing model had been bumping up against the rules around how controlled substances are supposed to be dispensed and monitored.

Some online providers responded by tightening their protocols, requiring in-person visits, working with state-licensed providers and reputable pharmacies. Some shut down entirely. Some moved to gray-zone international compounding arrangements, which raises a whole separate set of problems and is not a place anyone should be getting controlled substances from.

What’s still safe and reasonable

A board-certified psychiatrist or other appropriately credentialed prescriber, working with a reputable compounding pharmacy in your state, providing oral ketamine in the context of structured psychiatric care, with real follow-up and dose monitoring, is a defensible use. It’s how the medication was being used before the boom and how the better practitioners still use it. The difference between this and the online-app version is the actual clinical relationship doing actual clinical work.

Doses are typically in the 50mg to 200mg range, taken sublingually or buccally (held under the tongue or against the cheek for several minutes for absorption). Sessions can be spaced from a few times a week down to once a week to less frequent maintenance, depending on the patient and the response pattern. The cost of the actual medication is modest. The cost of the clinical care around it is what makes the difference between this being useful and this being a problem.

What’s grifty

Online providers who do a fifteen-minute Zoom intake, send a month’s supply, and “check in” via app without a real prescriber relationship. Especially if they’re selling escalating doses by default. Especially if they’re selling a “program” with no real clinical content beyond the dosing instructions and a vibe-curated soundtrack to listen to while dissociating.

Compounding pharmacies operating outside the state where the patient lives, in violation of state pharmacy laws, even if they’re technically licensed somewhere. The legal cover here is thinner than the marketing suggests, and the patient is the one left holding any bag that opens.

Anyone promising rapid permanent results without integration work, follow-up, or a real plan for when and how to taper. Ketamine works for the people it works for, but it’s not a one-and-done medication and anyone selling it as such is overselling.

The pattern this is about

Picture the kind of patient who ends up in trouble with the at-home programs: somebody who’s had treatment-resistant depression for years, tried multiple SSRIs and then Wellbutrin and SNRIs and maybe even supervised Spravato (which works but is a logistical pain to get to twice a week), and signs up for an online ketamine program because the at-home version is finally something accessible. The medication works for a few months. Then his original prescriber stops following him because the app is “handling it.” He’s refilling on his own through the app. The dose creeps up because more feels like more. He starts taking it more often than the original protocol called for. He’s using it more like a recreational tool than a treatment, and his depression has quietly drifted back to baseline because the underlying clinical work isn’t happening anymore.

The right move at that point is to bring him back into structured care, even if it means going back to the more inconvenient supervised setting that worked before. The lozenges had worked when supervised and stopped working when not, which says the variable that broke wasn’t the medication, it was the clinical relationship around it.

The variability was wide enough that nobody could defend the whole category honestly… some patients did remarkably well, some had bad experiences, and some developed dependence patterns.

The legal status in Oregon and Washington in 2026

Compounded ketamine is legal when prescribed by a licensed provider and dispensed by a licensed compounding pharmacy. Telehealth prescribing across state lines has tightened post-2024. Most reputable providers now require at least one in-person visit and prefer ongoing in-person care. Some practices in both states do this well. The handful of online operators still running app-only programs are operating in a legal gray zone that’s getting greyer.

If you’re considering compounded ketamine, look for a practice that does it as part of ongoing psychiatric care, not a standalone online service. The prescriber should be able to tell you what success looks like for you specifically, when you should be tapering or stopping, and what they’re monitoring for during treatment. If you can’t get a clear answer to those three questions, that’s information.

Where the autonomy stance lands

The decision to try compounded ketamine is the patient’s. My job is to be honest about what the medication does, where the legitimate uses end and the grifty uses begin, and what the difference between a real clinical relationship and an app-only “program” actually is. I’m a provider, not a parent. If somebody hears the trade-offs and wants to try compounded ketamine in the context of real care, that’s a reasonable decision. If somebody is already on an app-only version and the depression isn’t moving anymore, the honest take is that the program isn’t doing the clinical work and a switch back to a real provider relationship is what’s most likely to help, whether that’s Spravato or an oral protocol with actual oversight.

What’s nice to hear, because most of this post is about the things that went wrong with the boom, is that ketamine in any of its forms works for a chunk of patients where nothing else has worked. The medication itself is a real tool. The delivery model that turned it into a mail-order subscription with no real care attached is what created the problems, not the molecule.

Bottom line

Compounded ketamine has a real role in psychiatric care when handled properly. The boom of online unsupervised at-home ketamine programs was a mess that the FDA is still cleaning up. The medication itself is real. The way some of it was being delivered was not. If you’re considering it, find a real prescriber and a real clinical relationship, not an app. If you’re already on it through an online service, take a hard look at whether the care is actually clinical or just transactional. The difference matters more than the molecule does.

Sources

1. Daly EJ, Trivedi MH, Janik A, et al. Efficacy of Esketamine Nasal Spray Plus Oral Antidepressant Treatment for Relapse Prevention in Patients With Treatment-Resistant Depression. JAMA Psychiatry. 2019;76(9):893-903. PMID 31166571.
2. Andrade C. Oral Ketamine for Depression, 1: Pharmacologic Considerations and Clinical Evidence. J Clin Psychiatry. 2019;80(2):19f12820. PMID 30946543.
3. Wilkinson ST, Toprak M, Turner MS, Levine SP, Katz RB, Sanacora G. A Survey of the Clinical, Off-Label Use of Ketamine as a Treatment for Psychiatric Disorders. Am J Psychiatry. 2017;174(7):695-696. PMID 28669202.
4. Dean RL, Hurducas C, Hawton K, et al. Ketamine and other glutamate receptor modulators for depression in adults with unipolar major depressive disorder. Cochrane Database Syst Rev. 2021;9(9):CD011612. PMID 34510411.